The caret package (wiki) unifies the syntax for about 200 classification and regression models. However, picking a model is often a subjective task. This package aims to provide an example of how to iterate through all available models and quantify accuracy.
Simply grab from github.
library(devtools)
install_github("codychampion/Caret-Model-Evaluation")Once the package is installed make sure you have a complete installation of caret and ALL dependencies. This can be done using this command:
install.packages("caret",
repos = "http://cran.r-project.org",
dependencies = c("Depends", "Imports", "Suggests"))Use case is shown below:
library(datasets)
data <- data(iris)
target <- "Septal.Length"
modeleval(target, data, timeouttime = 60)Other important parameters are given in function discription.
The following is a graph showing optimal models for the iris dataset for Sepal length prediction as evaulesed by R2.
Some issues with this package are it will not work with the timeout option on Windows, Linux is needed.